International Journal of Science Engineering and Advance Technology, ISSN 2321-6905
IJSEAT, Vol 3, Issue 11, NOVEMBER - 2015
Dc-Bus Voltage Control With A Three-Phase Bidirectional Inverter
For Dc Distribution Systems
Venkata Manikanta Guttula1, Mr.B,D.S.Prasad2
PG Scholar, Pydah College of Engineering, Kakinada, AP, India.
2
Assistant Professor, Pydah College of Engineering, Kakinada, AP, India.
1
Abstract—In this paper a new energy management
system has been proposed for three-phase bidirectional
inverter with dc-bus voltage control. The advantage of
this bidirectional inverter is that it can operate in both
grid connection and rectification modes with power
factor correction. The proposed control system take into
account dc-bus capacitance and control dc-bus voltage to
track a linear relationship between the dc-bus voltage
and inverter inductor current. The inverter tunes the dcbus voltage every line cycle, which can reduce the
frequency of operation-mode change and current
distortion. With the increase in the demand of renewable
energy sources, placed important role in distributed
systems. In this paper battery storage system has been
proposed with DG system. Battery storage system will
give additional support to the DG system and also helps
in controlling the DC bus voltage. The simulation results
have been presented using MATLAB/SIMULINK
software. The simulation results validated for the
rectification mode with battery storage. This system will
improve the performance of the DG system.
inverter will buy power from ac grid, which is a
rectification, namely “rectification mode.” Moreover, since
in a dc distribution system, the dc-bus voltage is sensitive to
step load changes, the control of dc-bus voltage is more
critical than that in a grid-connected inverter system. In the
past studies, the voltage control based on gain scheduling
was presented, which use a droop concept to design proper
dc gain. Moreover, some attempts combing the gainscheduling with fuzzy control were also discussed, which
incorporate fuzzy control and adjust dc voltage reference to
balance power flow.
Index
Terms—Bidirectional
inverter,
capacitance
estimation, dc distribution system, dc-bus voltage control,
grid connection, rectification
I. INTRODUCTION
In an ac-bus system, the maximum power point
tracker (MPPT) will draw maximum power from
photovoltaic (PV) arrays and inject the power into ac bus or
ac grid through an inverter. For a typical offline application,
such as electronic ballast, personal computer, or variable
speed appliance, it usually has a rectifier, a power factor
corrector (PFC), and a dc/dc or dc/ac converter to supply the
loads. However, this ac system leads to high power
conversion loss. Thus, the concept of “dc grid” or “dc
distribution” system has been presented recently, and the
system configuration is illustrated in Fig. 1(b). It can be
calculated that the power conversion efficiency in the ac-bus
system is less than that in the dc-bus system about 8%. In
addition, the dc-bus system can also save one rectifier and
one PFC, saving component cost around 25%. Therefore,
the dc distribution system is feasible in renewable energy
applications.
The dc distribution system requires a dc-bus
voltage control to balance the power flow among PV panels,
dc loads, and ac grid. When the overall PV power is higher
than the dc load power, the bidirectional inverter, as shown
in Fig. 1(b), needs to sell power to ac grid, which is often
defined as grid-connection mode. On the contrary, the
www.ijseat.com
Fig. 1 Conceptual configurations of (a) an ac-bus system
and (b) a dc distribution system
Then, other dc-bus voltage controls, such as robust,
adaptive, and hybrid control to enhance system stability
were reported. However, a dc distribution system requires a
power management scheme to improve system availability;
that is, it is allowed to have a certain range of dc-bus voltage
variation for both grid-connection and rectification modes.
When the system is operated in grid-connection mode, it
needs a higher dc-bus voltage to prevent dramatic voltage
drop below the lower bound due to a step dc load increase.
On the contrary, it requires a lower dc-bus voltage to extend
the range of voltage swing in rectification mode. In the
literature, there are some power flow controls for dc
distribution system with constant-power loads, such as
general dc/dc converters, which behave with negative
dynamic impedance. The loads with negative dynamic
impedance will intend to regulate their input power under
dc-bus voltage variation.
Page 1031
International Journal of Science Engineering and Advance Technology, ISSN 2321-6905
IJSEAT, Vol 3, Issue 11, NOVEMBER - 2015
This effect can help the proposed control approach
to stabilize the dc-bus voltage regulation. Moreover, the
approaches for achieving fast dc-bus voltage dynamics were
focused on the systems without dc loads, while there is no
control for the bidirectional inverter systems with fast dc
load variation. In our previous research, a digital controlled
10-kVA 3φ bidirectional inverter with wide inductance
variation has been designed and implemented. Based on its
configuration, this paper presents two dc-bus voltage
regulation approaches, one line-cycle regulation approach
(OLCRA) and one-sixth line-cycle regulation approach
(OSLCRA), which are based on a linear power management
scheme. They are chosen by considering that the dc-bus
capacitance should be large enough (typically > 4000 μF) to
extend the hold-up time, and a short time interval does not
vary the dc-bus voltage significantly even under high power
change. Moreover, the OSLCRA is chosen for that the
control laws for each region are changed every one-sixth
line cycle, and the complexity of firmware programming can
be reduced. The OLCRA is enacting for light load change
while the OSLCRA is for heavy load change. The control
laws of the two approaches require the information of dcbus capacitance. It needs to determine the total capacitance
on the dc bus. In a dc distribution system, dc loads or dc
products having electrolytic capacitors are not desired to
prevent inrush current.
The capacitors are installed in the inverter or in a
separate box, and dc-bus capacitance estimation is only
required at the start-up of the system operation. Several
methods to estimate capacitance have been presented, where
a dc-bus capacitance estimation method for ac/dc/ac pulse
width modulation converter systems can determine the
capacitance online with voltage injection. In this paper, an
online capacitance estimation method with a capacitor precharging circuit at the system start-up is presented, which
does not need extra inverter input current feedbacks. The
proposed two control approaches can reduce the frequency
of operation-mode change and current distortion
significantly. In addition, they can improve the availability
of the dc distribution system without increasing dc-bus
capacitance.
II. DETERMINATION OF CAPACITOR SIZE AND
CAPACITANCE ESTIMATION
The proposed system design is based on the power
ratings of renewable sources, storage elements, dc load
requirements, and the bidirectional inverter. The size of
capacitors will affect the dc-bus voltage ripple and hold-up
time. Thus, it needs to determine its size first, and then
estimate the total capacitance online because of its tolerance.
A. Determination of Capacitor Size
In determining the capacitor size for the system,
two major aspects are considered: dc-bus voltage ripple and
energy-storage capability.
1) DC-Bus Voltage Ripple: By neglecting the ripple current
of the three-phase inductors, instantaneous output power
Pac from the inverter is given by
www.ijseat.com
Where VRN, VSN, and VTN are the amplitudes of the phase
voltages; IR, IS , and IT are their current amplitudes; ωl is
the line angular frequency; and θ is the phase between ac
currents and voltages. If the 3φ power source is unbalanced,
there will exist voltage ripple every one-twelfth line period,
Tl /12, and it can be derived from the following equation:
Where vDC MAX and vDC MIN are the peak and valley
values of the dc-bus voltage ripple, respectively, and Pavg is
the average power of the inverter input power. Assuming the
system is operated in the maximum power rating, 10kW,
two of the phase voltages have the maximum and minimum
unbalanced variations of 10%. A plot of dc-bus capacitance
CDC versus dc-bus voltage ripple is illustrated in Fig. 2. It
can be seen that when the capacitance is higher than 8 × 470
μF, the voltage ripple is lower than 1V. Therefore, we can
select PV capacitor Cpv = 12 × 470 μF, which is sufficient
enough to filter out the ripple effect from the unbalanced
grid voltages.
Fig. 2 Plot of dc-bus capacitance CDC versus dc-bus voltage
ripple vr under unbalanced grid voltages
Moreover, there is another voltage ripple effect
resulting from non-constant inductors. Assuming the threephase inductors are identical but with wide inductance
variation, the total energy variation will swing at six times
line frequency, corresponding to the six-region transitions.
The peak-to-peak voltage ripple can be then derived from
the following equation:
Fig. 3 shows a plot of CDC versus voltage ripple vr
under the maximum output power of 10 kW and the dc-bus
voltage of 380V. It can be observed that vr is lower than
0.25V even at only one 470-μF capacitor. Thus, the effect of
the voltage ripple due to wide inductance variation is not
Page 1032
International Journal of Science Engineering and Advance Technology, ISSN 2321-6905
IJSEAT, Vol 3, Issue 11, NOVEMBER - 2015
significant and can be ignored. However, the capacitor must
hold the dc-bus voltage during abrupt power changes, and it
will be determined according to energy-storage capability.
Where Vpv is the solar array voltage. The proposed
capacitance estimation is only working during the precharging time interval, because the dc distribution system
has themerit that all of dc-bus capacitors are installed at the
input terminal of the inverter or in a separate box, as shown
in Fig. 5. An illustration of the proposed estimation method
is shown in Fig. 6. During a sampling period ΔT, the
estimation equation in (4) can be rewritten as
Fig. 3 Plot of dc-bus capacitance CDC versus dc-bus voltage
ripple vr under a maximum output power of 10 kW
2) Energy-Storage Capability: Fig. 4 shows a plot of CDC
versus dc-bus voltage variation ΔvDC under a maximum
power change of 10kW and within one-sixth line cycle. For
power supply availability, the voltage variation must be
lower than 20V (400 –380V or 380–360 V) within one-sixth
line cycle, and 12× 470-μF capacitance can be selected to fit
the appropriate range of dc-bus voltage variation, 10–15V.
This set of capacitors was used in the experiment.
Fig. 4 Plot of dc-bus capacitance CDC versus dc-bus voltage
variation ΔvDC under a 10-kW abrupt change and within
one-sixth line cycle
The inverter microcontrollers will feedback the voltages,
vDC and vpv, to determine the dc-bus capacitance from (6).
Moreover, the inverter controller will estimate the
capacitance automatically when the auxiliary power output
is available and the CPU is working. At this moment, the
CPU’s timer starts to count.
Fig. 5 Configuration of the proposed dc distribution system
with a capacitor pre-charging circuit
B. Capacitance Estimation
The proposed online capacitance estimation
method uses the relationship between dc-bus voltage
variation and capacitor current injection, and is based on the
capacitor pre-charging circuit. The overall system
configuration with a pre-charging circuit is shown in Fig. 5,
where the circuit includes a resistor R and a relay (normally
closed). When any of the three solar arrays generates power,
it will charge the dc-bus capacitor through the resistor. The
capacitance CDC can be then determined as
Where ΔvDC is the dc-bus voltage variation during time
interval ΔT, and iR is the current through resistor R. Since
there is no feedback from resistor current iR, it is determined
as
www.ijseat.com
Fig. 6 Illustration of the proposed capacitance estimation
method
Page 1033
International Journal of Science Engineering and Advance Technology, ISSN 2321-6905
IJSEAT, Vol 3, Issue 11, NOVEMBER - 2015
Although the dc-bus voltage variation ΔvDC will
change with different start-up times, it does not deteriorate
in the estimating results. In the estimation tests, the
sampling period ΔT was 10ms, resistor R = 200 Ω was
selected, and four different dc-bus capacitances, 8 × 470, 10
× 470, 12 × 470, and 16 × 470 μF, were, respectively,
implemented. A comparison between the estimated and
measured data for the different dc-bus capacitances is listed
in Table I. The estimation errors are all within 1%, which
does not penalize the accuracy of the dc-bus voltage
regulation.
TABLE I
COMPARISON BETWEEN MEASURED AND
ESTIMATED CAPACITANCES
Fig. 7 Diagram of the proposed three-phase bidirectional
inverter
C. DC-BUS VOLTAGE CONTROL
A diagram of the discussed three-phase
bidirectional inverter is shown in Fig. 7. It can fit to both
delta-connected and Yconnected ac grid. In the designed
prototype, Renesas microchip RX62T is adopted for
realizing the system controller, which has 1.65 MIPS and
includes floating-point calculation and division. By
considering wide inductance variation, the inverter can be
operated stably, especially in high-current applications.
Additionally, the system requires dc-bus voltage control
schemes for balancing power flow. It includes linear power
management scheme, one line-cycle regulation approach,
and one-sixth line cycle regulation approach. A stability
analysis is presented to support the proposed regulation
approaches.
1. Linear Power Management Scheme
In a system design, the maximum PV power and
the maximum dc load power will not exceed the inverter
capability. In our developed system, both of the two
maximum powers are 10kW. Fig. 8 shows an illustration of
the proposed linear power management scheme for
achieving a linear relationship between inductor current iL
and dc-bus voltage vDC. The operating range of the dc-bus
voltage is 380±20V. When the inverter is operated in gridconnection mode (selling power), the operating range is
from 380 to 400V, which ensures an enough voltage level to
accommodate abrupt dc load increase. On the other hand,
when the inverter is operated in rectification mode (buying
power), a lower dc-bus voltage stands for a higher load
power. Therefore, in rectification mode, the system does not
have extra high enough dc load power to pull the dc-bus
voltage below the lower bound. In short, in grid-connection
mode, dramatic voltage drop is the major concern when
loads are connected to the dc bus, while in rectification
mode, sudden voltage jump up is another concern when
loads are disconnected from the dc bus.
www.ijseat.com
Fig. 8 also shows the locus of the dc-bus voltage
regulation sequence in grid-connection mode. At time t0, the
inverter stays at operating point (vDC(n – 1), IA(n – 1)) on
the load line, where IA is the adjustable current command
which is tuned every line cycle with the OLCRA. The
operating point will move away from the load line to point
(vDC(n), IA(n – 1)) at t1 when there exists power imbalance.
Then, the controller will update current command IA(n – 1)
with IA(n) at the beginning of the nth line cycle, which will
regulate dc-bus voltage to a new set-point voltage vDC(n +
1) according to the linear power management scheme. When
the inverter reaches operating point (vDC(n+1), IA(n)) at t2 ,
the controller will change current command to IA(n + 1),
maintaining dc-bus voltage vDC for a new power balance.
The system will be operated in the new steady state after
time t3. The proposed voltage regulation approaches are
based on the linear power management scheme and
presented in the following section.
Fig. 8 Illustration of a linear power management scheme for
achieving a linear relationship between inductor current iL
and dc-bus voltage vDC
2 One Line-Cycle Regulation Approach
Fig. 9 shows a diagram for illustrating the OLCRA
when power imbalance is occurring. In Fig. 9, power
imbalance happens at time tx 1 , and dc-bus voltage, for
Page 1034
International Journal of Science Engineering and Advance Technology, ISSN 2321-6905
IJSEAT, Vol 3, Issue 11, NOVEMBER - 2015
instance, will increase from vDC(n – 1) to vDC(n). At time
tn, dc-bus capacitor current variation ΔiC can be determined
as
inverter system with even 10% unbalanced three-phase
voltage sources, the voltage ripple can be ignored in the
control law derivation.
Fig. 10 illustrates the current tuning process of the proposed
OSLCRA. With the OSLCRA, the inverter will update
current command IA at the beginning of each one-sixth linecycle to drive dc-bus voltage to set-point voltage vDC(n + 1)
at time tn+1. Thus, according to a dc-bus voltage variation,
the current command control law can be derived as follows:
Fig. 9 Illustration of the OLCRA to dc-bus voltage
regulation for the dc distribution system with power
imbalance
However, time tx 1 is unpredictable, and an initial guess is
made to tn− 1 ; thus, new current command IA(n) for time
interval [tn , tn+1] can be determined based on previous
current IA(n – 1) as
where Tl is a line period, KDA is a mapping ratio
between dc input current and ac output current, which is
equal to vDC/VAC (VAC is the RMS value of ac grid
voltage, typically 220 V), and the initial current command
IA(0) is equal to zero at the system start-up. Once the system
gets started, current command IA(n – 1) will be updated
cycle by cycle. Voltage vDC(n + 1) is the set-point voltage,
which can be derived based on the relationship shown in
Fig. 8, as follows:
In the above equations, IA(n) is always used for
regulating dc-bus voltage to set-point voltage vDC(n + 1)
according to the control diagrams shown in Figs. 8 and 9.
The inverter controller will update current command IA(n)
cycle by cycle, that is, IA(i) will be loaded to IA(I – 1) after
the new current command has been determined at the
beginning of a line cycle. In Fig. 9, if another power
imbalance occurs at tx 2 (curve 2), the inverter will tune
current command IA continuously to follow curve 2 and
reach another set-point.
3. One-Sixth Line-Cycle Regulation Approach
If dc loads change abruptly, the OLCRA cannot
regulate dc bus voltage immediately, and it needs a fast
dynamic current control to balance the power flow. In a
three-phase inverter system, the fast regulation interval is
selected to be one-sixth line-cycle according to each zerocrossing point of the three phase line currents. Since the
voltage ripple on the dc-bus is insignificant in a three-phase
www.ijseat.com
Voltage variation Δvi includes two portions: one is the
variation ΔvDC(T ) between two consecutive one-sixth
cycles, and the other is the average difference ΔvDC(A)
between vDC(n+1) and the current operating point:
And
Fig. 10 Illustration of the OSLCRA to dc-bus voltage
regulation for the system under power imbalance
The variation ΔvDC(T ) is caused by abrupt power
imbalance, of which the dc-bus voltage will deviate from the
load line. By adding this variation to the current command
in one-sixth cycle, the inverter can balance power flow
substantially, but dc-bus voltage vDC still does not reach its
set-point yet. Thus, the variation ΔvDC(A) is used for
regulating vDC to voltage vDC(n + 1) which has been
determined at the beginning of the nth line cycle, as shown
in (9). With the compensation of these two portions, the dcbus voltage will come back to the load line after power
imbalance occurs at ty 1 (curve 2) or ty 2 (curve 3), as
shown in Fig. 10. It can be observed that the inverter
controller updates current command IA(n + 1/6) at tn + Tl /6
after dc-bus voltage drops abruptly at ty 1 (curve 2).
However, the inverter controller will determine a wrong
current command due to the unpredictable time tY 1 , and
Page 1035
International Journal of Science Engineering and Advance Technology, ISSN 2321-6905
IJSEAT, Vol 3, Issue 11, NOVEMBER - 2015
the dc-bus voltage will not be regulated to a correct voltage
value until next one-sixth line cycle, tn + 2Tl/6. If another
power imbalance occurs at ty 2 (curve 3), the same
regulation process will be conducted again. That is, the
inverter will use a correct current command IA(n + 3/6) to
regulate the dc-bus voltage to a set-point voltage.
Furthermore, current command IA(n) must be modified for
determining command IA(n + 1) at the beginning of a new
cycle due to the one-sixth line-cycle command change, and
it is just equal to the average value of all current commands
within the time interval of [tn , tn+1]. The determination of
current command IA(n) can be expressed as
III. SIMULATION RESULTS
Fig. 12 Measured waveforms of the three-phase inductor
currents and dc-bus voltage vDC for dc load variation (a)
from no load to 3kW and (b) from 3 to 1.5 kW in
rectification mode with BSS
Fig 11 simulation diagram of proposed system
Fig 13 measured waveforms of BSS
Fig. 11.Measured waveforms of the three-phase inductor
currents and dc-bus voltage vDC for dc load variation (a)
from no load to 800Wand (b) from 800W to 2 kW in
rectification mode with BSS
www.ijseat.com
CONCLUSION
A dc-bus voltage control for three-phase
bidirectional inverters in dc distribution systems has been
presented. The linear power management scheme including
both grid-connection and rectification modes have been
described in detail. In the paper, the one line-cycle
regulation approach based on the linear power management
scheme has been proposed for tuning current command
cycle by cycle to regulate the dc-bus voltage according to
the load line. For an abrupt voltage change, the one-sixth
line cycle regulation approach has been also proposed to
prevent over/under dc-bus voltage fault, ensuring system
availability. Since both of the approaches are dependent on
the parameter of dc-bus capacitance, a determination of
capacitor size and a capacitance estimation method have
been discussed according to the aspects of dc-bus voltage
ripple and energy-storage capability.
Page 1036
International Journal of Science Engineering and Advance Technology, ISSN 2321-6905
IJSEAT, Vol 3, Issue 11, NOVEMBER - 2015
REFERENCES
[1] T.-F. Wu, K.-H. Sun, C.-L. Kuo, and C.-H. Chang,
“Predictive current controlled 5-kW single-phase
bidirectional inverter with wide inductance variation for dcmicrogrid applications,” IEEE Trans. Power Electron., vol.
25, no. 12, pp. 3076–3084, Dec. 2010.
[2] L. Xu and D. Chen, “Control and design of a DC
microgrid with variable generation and energy storage,”
IEEE Trans. Power Del., vol. 26, no. 4, pp. 2513–2522, Oct.
2011.
[3] Z.-H. Ye, D. Boroyevich, K. Xing, and F.-C. Lee,
“Design of parallel sources in DC distributed power systems
by using gain-scheduling technique,” in Proc. IEEE Power
Electron. Spec. Conf., Aug. 1999, pp. 161–165.
[4] Y. Ito, Y. Zhongqing, and H. Akagi, “DC microgrid
based distribution power generation system,” in Proc. IEEE
Int. Power Electron. Motion Control Conf., Aug. 2004, pp.
1740–1745.
[5] J. M. Guerrero, J. C. Vasquez, J. Matas, L. G. D. Vicuna,
and M. Castilla, “Hierarchical control of droop-controlled
AC and DC microgrid—A general approach toward
standardization,” IEEE Trans. Ind. Electron., vol. 58, no. 1,
pp. 158–172, Jan. 2011.
[6] H. Kakigano, A. Nishino, and T. Ise, “Distributed
voltage control for DC microgrid with fuzzy control and
gain-scheduling control,” in Proc. IEEE Int. Conf. Power
Electron., 2011, pp. 256–263.
[7] H.Kakigano,Y. Miura, and T. Ise, “Low-voltage bipolartype dc microgrid for super high quality distribution,” IEEE
Trans. Power Electron., vol. 25, no. 12, pp. 3066–3075,
Dec. 2010.
[8] J.-S. Park, J.-K. Choi, B.-G. Gu, I.-S. Jung, E.-C. Lee,
and K.-S. Ahn, “Robust DC-Link voltage control scheme for
photovoltaic power generation system PCS,” in Proc. IEEE
Int. Telecommun. Energy Conf., Oct. 2009, pp. 1–4.
[9] D. Salomonsson, L. Soder, and A. Sannino, “An
adaptive control system for a DC microgrid for data
centers,” IEEE Trans. Ind. Appl., vol. 44, no. 6, pp. 1910–
1917, Nov./Dec. 2008.
[10] J.-S. Park, J.-K. Choi, B.-G. Gu, I.-S. Jung, E.-C. Lee,
and K.-S. Ahn, “A hybrid renewable DC microgrid voltage
control,” in Proc. IEEE Int. Power Electron. Motion Control
Conf., May 2009, pp. 725–729.
[11] Z. Li, T. Wu, X. Yan, K. Sun, and J. M. Gurrero,
“Power control of DC microgrid using DC bus signaling,” in
Proc. IEEE Appl. Power Electron. Conf., Mar. 2011, pp.
1926–1932.
[12] A. Engler, “Applicability of droops in low voltage
grids,” Int. J. Distrib. Energy Res., vol. 1, no. 1, pp. 1–5,
Jan. 2005.
[13] J. A. Restrepo, J. M. Aller, A. Bueno, J. C. Viola, A.
Berzoy, R. Harley, and T. G. Habetler, “Direct power
control of a dual converter operating as a synchronous
rectifier,” IEEE Trans. Power Electron., vol. 26, no. 5, pp.
1410–1417, May 2011.
[14] M. G. Molina and P. E. Mercado, “Power flow
stabilization and control of microgrid with wind generation
by superconducting magnetic energy storage,” IEEE Trans.
Power Electron., vol. 26, no. 3, pp. 910–922, Mar. 2011.
[15] Y. C. Chang and C. M. Liaw, “Establishment of a
switched-reluctance generator-based common DC microgrid
www.ijseat.com
system,” IEEE Trans. Power Electron., vol. 26, no. 9, pp.
2512–2527, Sep. 2011.
[16] N. Hur, J. Jung, and K. Nam, “A fast dynamic DC-link
power-balance scheme for a PWM converter-inverter
system,” IEEE Trans. Ind. Electron., vol. 48, no. 4, pp. 794–
1803, Aug. 2001.
[17] J. Yao, H. Li, Y. Liao, and Z. Chen, “An improved
control strategy of limiting the DC-link voltage fluctuation
for a doubly fed induction wind generator,” IEEE Trans.
Power Electron., vol. 23, no. 3, pp. 1205–1213, May 2008.
[18] T.-F. Wu, L.-C. Lin, C.-H. Chang, Y.-L. Lin, and Y.-C.
Chang, “Current improvement for a3 bi-directional inverter
with wide inductance variation,” in Proc. IEEE ECCE Asia,
May/Jun. 2011, pp. 1777–1784.
[19] E. Aeloiza, J.-H. Kim, P. Enjeti, and P. Ruminot, “A
real time method to estimate electrolytic capacitor condition
in PWM adjustable speed drives and uninterruptible power
supplies,” in Proc. IEEE Power Electron. Spec. Conf., Jun.
2005, pp. 2867–2872.
[20] A.-G. Abo-Khalil and D.-C. Lee, “DC-link capacitance
estimation in AC/DC/AC PWM converters using voltage
injection,” IEEE Trans. Ind. Appl., vol. 44, no. 5, pp. 1631–
1637, Sep. 2008.
[21] A. J. Roscoe, S. J. Finney, and G. M. Burt, “Tradeoffs
between AC power quality and DC bus ripple for 3-Phase 3wire inverter-connected devices within microgrids,” IEEE
Trans. Power Electron., vol. 26, no. 3, pp. 674– 688, Mar.
2011.
[22] G. Petrone, G. Spagnuolo, and M. Vitelli,
“Optimization of perturb and observe maximum power point
tracking method,” IEEE Trans. Power Electron., vol. 20, no.
4, pp. 963–973, Jul. 2005.
[23] A. Khaligh, C. H. Rivetta, and G. A. Williamson,
“Constant power loads and negative impedance instability in
automotive systems: Definition, modeling, stability, and
control of power electronic converters and motor drives,”
IEEE Trans. Veh. Technol., vol. 55, no. 4, pp. 1112–1125,
Jul. 2006.
[24] P. Liutanakul, A. Awan, S. Pierfederici, B. NahidMobarakey, and F. Meibody-Tabar, “Linear stabilization of
a DC bus supplying a constantpower load,” IEEE Trans.
Power Electron., vol. 25, no. 2, pp. 475–488, Feb. 2010.
[25] A. Kwasinski and C. N. Onwuchekwa, “Dynamic
behavior and stabilization of DC microgrids with
instantaneous constant-power loads,” IEEE Trans. Power
Electron., vol. 26, no. 3, pp. 822–834, Mar. 2011.
Page 1037